Electronic and interface properties of polyfluorene films on GaN for hybrid
optoelectronic applications
G. Itskos, X. Xristodoulou, E. Iliopoulos, S. Ladas, S. Kennou et al.
Citation: Appl. Phys. Lett. 102, 063303 (2013); doi: 10.1063/1.4792211
View online: http://dx.doi.org/10.1063/1.4792211
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v102/i6
Published by the AIP Publishing LLC.
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
APPLIED PHYSICS LETTERS 102, 063303 (2013)
Electronic and interface properties of polyfluorene films on GaN for hybrid
optoelectronic applications
G. Itskos,1,a) X. Xristodoulou,1 E. Iliopoulos,2,3 S. Ladas,4 S. Kennou,4 M. Neophytou,5
and S. Choulis5
1
Experimental Condensed Matter Physics Laboratory, Physics Department, University of Cyprus,
Nicosia 1678, Cyprus
2
Institute of Electronic Structure and Laser (IESL), Foundation for Research and Technology-Hellas
(FORTH), 71110 Heraklion, Greece
3
Physics Department, University of Crete, 71003 Heraklion, Greece
4
Department of Chemical Engineering, University of Patras, Patras GR-26504, Greece
5
Molecular Electronics and Photonics Research Unit, Department of Mechanical Engineering and
Materials Science and Engineering, Cyprus University of Technology, Limassol 3603, Cyprus
(Received 21 November 2012; accepted 31 January 2013; published online 13 February 2013)
Electronic and interface properties of spin-coated poly(9,9-dioctylfluorenyl-2,7-diyl) (PFO)
films on GaN have been investigated in terms of their potential for optoelectronic applications. The
PFO/GaN interface was studied by photoemission spectroscopy showing a type-II energy
alignment with band offsets suitable for efficient photocurrent generation. The light harvesting
potential is further supported by fluorescence experiments that show evidence of photo-induced
electron transfer from PFO to GaN. The impact of polymer film thickness was probed using
emission anisotropy and ellipsometry, indicating the presence of an ordered planar phase of PFO.
The study has implications to hybrid optoelectronic devices employing the two important
C 2013 American Institute of Physics. [http://dx.doi.org/10.1063/1.4792211]
materials. V
Heterogeneous structures of organic and inorganic semiconductors can employ the favorable properties of each class
of materials to offer improved functionalities to optoelectronic devices.1 The use of solution processing methods to
deposit the organic component on the top surface of an inorganic semiconductor wafer adds to the ease of fabrication of
planar hybrid structures and allows the monolithic incorporation of different semiconductors that are not necessarily lattice matched. This approach creates enormous possibilities
of combining organic and epitaxial inorganic semiconductors
into hybrid optoelectronic devices but also imposes the need
for fundamental studies of the intermaterial interactions and
the nature and disorder of the hybrid interface(s).
Assemblies of nitrides and organics for optoelectronic
applications have been under investigation since the early
stage development of GaN.2,3 Efforts have mostly concentrated on light emission based on downconversion by the
nitride to achieve white light,2,4–6 multi-color microarrays,7,8 and even polymer lasing.9 Nitride-organic light
harvesting have been more recently under investigation with
demonstrations of photosensors10,11 and solar cells.12 Some
of the most promising studies combined (Ga,In)N structures
with polyfluorenes known for their high absorption coefficients and fluorescence quantum efficiencies.13,14 Planar
structures of the aforementioned materials have been
employed to demonstrate efficient emission across the entire
visible, coupled radiatively via downconversion7,8 or nonradiatively via Forster resonant energy transfer (FRET).15–17
In both cases, intermaterial interactions are expected to
be strongly affected by the heterointerface. Interfacial disorder and band alignment are also critical parameters for the
a)
Author to whom correspondence should be addressed. Electronic mail:
itskos@ucy.ac.cy.
0003-6951/2013/102(6)/063303/5/$30.00
performance of potential Polyfluorene/GaN junction devices
such as those of Refs. 10–12 yet no relevant studies exist in
the literature. Furthermore, some of the aforementioned
structures employ surface quantum wells (QWs)18 and ultrathin organic films, both of which have been scarcely
employed in devices before and thus are rather unexplored.
We report on the investigation of spin-coated ultrathin
to thin (5–55 nm) films of the prototype polyfluorene
homopolymer poly(9,9-dioctylfluorenyl-2,7-diyl) (PFO) deposited on GaN wafers and control fused silica and sapphire
(Al2O3) substrates. A combination of optical spectroscopy,
namely steady-state and time-resolved photoluminescence (PL)
emission, excitation (PLE) and anisotropy experiments,
and surface techniques such as X-ray and ultraviolet photoemission and ellipsometry have been employed to probe the electronic and interface properties of the hybrid structures.
Poly(9,9-dioctylfluorenyl-2,7-diyl) (PFO) end capped
with N,N-Bis(4-methyl-phenyl)-4-aniline was purchased from
American Dye Source Inc (ADS329BE). The material was
diluted in a 1 mg/ml concentration in chlorobenzene. The solution was left stirring at 80 C for more than one hour and was
spin-coated on polar [0001] c-plane GaN, c-plane Al2O3 and
fused silica (spectrosil B) substrates that have been cleaned by
acetone and isopropyl alcohol. Apart from the cleaning, no
other chemical treatment of the substrate surfaces was carried
out. Rotational speeds of 3000–5000 rpm were used, followed
by a 6000 rpm drying step, to achieve polymer films in the
range of 5 to 55 nm. Film thicknesses were measured with a
Dectak Profilometer with accuracy of 62 nm.
To probe the surface and interface characteristics of
PFO on (0001) GaN, two samples, a reference pristine GaN
and a hybrid structure composed of a 5 nm thin PFO on GaN
were introduced to ultra-high vacuum conditions and characterized by X-ray and ultraviolet photoemission spectroscopy
102, 063303-1
C 2013 American Institute of Physics
V
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
063303-2
Itskos et al.
(XPS/UPS). Unmonochromatized Mg Ka line at 1253.6 eV and
a hemispherical analyzer with multichannel detector, at a pass
energy of 100 eV, were used in XPS measurements. For the
UPS measurements, the HeI (21.22 eV) excitation line was
employed. A negative bias of 12.28 V was applied to the sample
during UPS measurements in order to separate sample and analyzer high binding energy cut-off and estimate the absolute
work function value from the UPS spectra.
The XPS spectra show that only C, O, Ga, and N are present in both samples in the analysis depth (about 10 nm). In the
reference sample C and O are due to atmosphere contamination.
In the PFO/GaN sample, Ga and N are coming from the
substrate since the PFO layer is thinner than the analysis depth.
Oxygen is also due to contamination. Fig. 1 shows the UPS valence band spectrum of the reference GaN (top) and the hybrid
PFO/GaN (bottom) sample. For the control GaN sample, the
valence band (VB) edge obtained from the right part of the spectrum is determined at 2.1 eV from the Fermi level. The work
function obtained from the high binding energy cutoff was estimated at 4.4 eV. From these values, an ionization energy
(IE ¼ (EVac EVB) of 6.5 6 0.1 eV is estimated, where EVac is
the vacuum energy. Considering an energy gap of Eg ¼ 3.4 eV
at room temperature, the electron affinity (EA ¼ EVac ECB) is
found equal to 3.1 6 0.1 eV, in good agreement with values
reported in the literature.19,20 The energy state values measured
indicate an n-type character of the GaN substrate.
The bottom part of Fig. 1 shows the valence band spectrum of the PFO/GaN sample. As the analysis depth of the
UPS is of the order of 3 nm, the spectrum is mainly related to
the PFO valence band where contributions from the phenyl
back bone and the alkyl side chains exist. The HOMO onset is
estimated at 1.5 eV from the Fermi level and the high binding
energy cutoff at 4.1 eV. The peaks labeled A and B at 2 eV
FIG. 1. UPS spectra of: GaN control (top) and PFO/GaN (bottom) samples.
In magnification, the low energy part of the spectrum where the valence
band edge is obtained from (right) and the high binding energy cutoff where
the work function is obtained from (left), are shown.
Appl. Phys. Lett. 102, 063303 (2013)
and 4 eV in the figure are attributed to the localized p states
of the polymer backbone. Using the HOMO and energy cutoff
values, the ionization potential is estimated at 5.6 6 0.1 eV, in
agreement with the literature.21 From the experimental data,
the energy level diagram at the interface is drawn in Fig. 2.
The measurements confirm a type-II alignment with
CB/LUMO and VB/HOMO band offsets of 0.4 and 0.6 eV,
respectively. The CB/LUMO offset is comparable and slightly
higher to measured values of the PFO exciton binding energy
(0.3 eV)22 indicating an energy level alignment that can promote efficient photocurrent generation, as it allows for interfacial exciton dissociation via an electron transfer process to
GaN while potentially minimizing electron thermal losses in
the acceptor (GaN) material. Considering a Fermi level alignment of the two materials, evidenced by the absence of XPS
peaks shifts in the two samples and supported by previous
studies of the PFO interface with Au, ITO, and PEDOT:PSS
(Ref. 23), the measurements indicate the formation of a small
interface dipole of 0.3 6 0.1 eV.
Steady-state and time-resolved PL emission and PLE
were employed to investigate the influence of thickness and
substrate type on the spectral and temporal characteristics of
the PFO fluorescence. Steady-state PL was carried out using
a 0.75 m spectrometer equipped with a CCD camera, excited
either by the 355 nm line of the triple harmonic of a nanosecond pulsed Nd:YAG laser or a chopped (20 Hz) 375 nm laser
diode. PLE was excited by an ozone-free 450 W Xenon lamp
and monitored by a spectrophotometer equipped with a photomultiplier tube. Time-resolved PL was measured by a
0.35 m spectrometer-based, time-correlated single-photon
counting (TCSPC) setup using a 375 nm ps laser diode. The
system exhibits a time-resolution of 50 ps after reconvolution with the instrument response function.
Normalized PLE spectra (top) of identical 15 nm PFO
films on GaN and Al2O3 are displayed on the left hand side
of Fig. 3 (top). The PLE spectrum of the latter film contains
a feature peaked at 400 nm, typical of PFO p-p* transitions
and qualitatively matches the film absorption spectrum. The
excitation spectra of the GaN deposited film on the other
hand, contain a structured, broad feature extending further
into UV due to strong emissive contributions from the GaN
band-edge states. This is supported by fluorescence experiments performed using photo-excitation above the GaN
energy gap at 355 nm, that show significant enhancement of
the polymer emission in GaN relative to films on sapphire or
glass, due to radiative pumping by the underlying GaN layer.
FIG. 2. Energy level diagram of the PFO/GaN interface obtained by UPS
studies of pristine GaN and PFO/GaN samples.
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
063303-3
Itskos et al.
FIG. 3. PL (right) and PLE (left) spectra of 15 nm PFO films on GaN and
Al2O3 (top). Normalized PL decays monitoring the 0-0 peak of PFO of the
aforementioned samples in a logarithmic scale (bottom). The instrument
response is also displayed.
However, the situation reverses under photoexcitation at
375 nm, below the GaN band-edge, where the PL emission
intensity of sapphire deposited films is systematically higher
than those deposited on GaN. The comparative PL spectra of
the 15 nm films on the two types of substrates are shown on
the right hand-side of Fig. 3 (top). Both films exhibit the
characteristic singlet exciton PFO vibronic progression with
the GaN deposited film showing a high intensity ratio of the
0-0 to the higher vibronic modes for reasons discussed later
in the manuscript. Of interest here is that the integrated emission of PFO on GaN is quenched by a factor of 2 relative
to that of the film on sapphire. Overall, a consistent quenching of the PFO/GaN integrated emission by factors of 1.25
to 3 compared to films of equal thicknesses on sapphire,
for thicknesses up to 55 nm, has been observed.
The quenching is also present at the dynamics of the
polymer exciton emission. Fig. 3 (bottom) displays PL decays
of the two 15 nm films on GaN and Al2O3 obtained by photoexcitation at 375 nm. The displayed traces are obtained when
the temporal evolution of the 0-0 peak of the PFO emission
was monitored. The two decays are to a good approximation
mono-exponentials with time constants of 445 ps and
375 ps for the film on Al2O3 and GaN, respectively, i.e., a
faster decay by 20% is measured in the latter relative to the
former film. Overall quenching of the PFO exciton lifetime in
GaN deposited films relative to films on sapphire, by factors
of 10%–25% were measured, with higher quenching factors
generally observed for thinner film comparisons.
The systematic appearance of steady-state and timeresolved fluorescence quenching in the films on GaN relative
to those on Al2O3 are indicative of the presence of an
additional exciton relaxation channel in the former films. We
Appl. Phys. Lett. 102, 063303 (2013)
attribute the channel to photo-induced electron transfer processes from the PFO to GaN across the heterointerface. Such
processes are energetically allowed by the measured band
offsets as explained above, and are consistent with the observation of larger TR-PL quenching for thinner PFO films.
Thinner films allow for a higher exciton dissociation probability on the PFO/GaN interface as the Frenkel exciton diffusion length of 5 nm (Ref. 24) compares more favorably to
the film thicknesses. Indicative values of the electron transfer
times can be obtained by computing the decay rates of the
films on GaN and Al2O3. In the example of the Fig. 3, the
decay rates of the films on GaN and Al2O3 are 2.67 ns1
and 2.25 ns1, respectively. The electron transfer rate can
then be estimated by the difference of the two decay rates,
namely, 2.67–2.25 ns1 ¼ 0.42 ns1 from which an electron
transfer time of 2.4 ns is obtained. Overall electron transfer
times from PFO to GaN of 2–10 ns were obtained in the
studied films using the described comparative TR-PL
method. Higher electron transfer rates for polyfluorene/GaN
device applications may be obtained by further studies that
can investigate in detail the interfacial disorder and the influence of nitride surface termination, i.e., III-rich or N-rich
growth and polarity, i.e., polar versus non-polar substrates,11
on the GaN/PFO energy level alignment.
The impact of film thickness on the photophysics of
PFO films on GaN, Al2O3, and glass substrates was also
studied. Variable angle spectroscopic ellipsometry (VASE)
measurements on thin (15 nm) polyfluorene (PFO) films on
fused silica (spectrosil B) substrates were performed. The
measurements were taken in the spectral region of 300–
1000 nm, for a range of incidence angles from 50 to 80 .
The data were analyzed using a simple bilayer optical model
(PFO/Glass), with the PFO dielectric function modeled by
generalized oscillators25 to ensure Kramers-Kronig consistency. The results of the PFO deduced dielectric function,
expressed as refractive index n and extinction coefficient k,
are shown in Fig. 4 (bottom).
The model assumes isotropic optical properties of the film
taking into account the amorphous nature of the substrate. The
assumption is supported by previous studies of PFO films on
glass26 and PL anisotropy measurements presented later in the
manuscript that show small values of emission anisotropy in
PFO films deposited in fused silica. In the top of Fig. 4, the
measured ellipsometric angles for two incident angles are
shown and compared to the simulated ones (red lines) based
on the deduced PFO dielectric function. Compared to similar
VASE measurements on thicker PFO films26 (>100 nm), the
ultrathin films of our studies exhibit increased oscillator
strengths of the 0-0 vibronic and slightly higher values of the
refractive index n and extinction coefficient k.
Support of the ellipsometry findings is provided by PL
experiments. Comparative PL measurements show that thin
films of thicknesses typically less than 25–30 nm deposited
on either GaN or Al2O3 exhibited unusually strong emission
and higher intensity ratios of the 0-0 to the higher vibronic
modes. To further investigate the effect, a TR-PL study was
carried out. The results of such a study for films on GaN substrates when monitoring the 0-0 vibronic mode are displayed
on Fig. 5. A correlation between PFO exciton lifetime and
film thickness is evident with thinner films showing a faster
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
063303-4
Itskos et al.
Appl. Phys. Lett. 102, 063303 (2013)
FIG. 4. Variable angle ellipsometry of a 15 nm PFO spin-coated on glass:
Example of the fitted model to experimental data for incidence angles of 60
and 70 degrees (top). Derived complex dielectric function of PFO, expressed
as refractive index n and extinction coefficient k (bottom).
decay. In particular, as the film thickness decreases from 55
to 10 nm, the decay constants decrease from 400 to 350
ps, respectively. A similar trend of decay lifetime versus
thickness was also observed on PFO films on sapphire. The
faster decays of thinner films can be predominantly attributed to higher fluorescence rates consistent with the strong
emission they exhibit in the steady-state spectra.
To explore the origin of the effect, we performed timeresolved measurements of the photo-induced anisotropy on
the films. The linearly polarized line of the pulsed 375 nm
laser was used as the excitation, while the linear polarization
of the emitted luminescence was analyzed by a GlanThomson polarizer, placed before the detecting spectrometer.
The geometry employed is graphically displayed at the top of
Fig. 6. The sample was excited at nearly normal incidence
FIG. 6. IVV* (red) and IVH* (blue) decays and emission anisotropy coefficient r(t) (green) for a 10 nm PFO film on GaN and r(t) for a thicker 25 nm
film on GaN (orange) (top). Anisotropy coefficient at t ¼ 0 of different thickness PFO films on GaN and Al2O3 (bottom).
and the side-emission in the vicinity of the 0-0 PFO vibronic
peak was spectrally and temporally resolved and detected.
The polarization of the excitation and emission was selected
between the vertical and horizontal directions defined on the
display of Fig. 6. Experimentally four decays per sample
were measured, defined by the indexes IVV, IVH, IHV, and
IHH, where the first index denotes the polarization of the excitation and the second the polarization of the emission.
The emission anisotropy is typically quantified by the
anisotropy coefficient defined as
r¼
FIG. 5. Normalized PL decays of PFO films of different thicknesses
(10–55 nm) on GaN obtained at the 0-0 vibronic peak of PFO.
IVV GIVH
;
IVV þ 2 GIVH
(1)
where G ¼ IIHV
is the correction factor that accounts for the
HH
different sensitivity of the detection system to vertical and
horizontal polarizations of light.
The temporal evolution of the emission anisotropy coefficient r(t) was recorded for four film thicknesses of 10, 15,
25, and 55 nm on GaN and sapphire, a total of 8 samples.
Fig. 6 (middle) shows the instrument corrected IVV* and IVH*
decays (red and blue curves, respectively) and the calculated
emission anisotropy coefficient r(t) (green curve) for the
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
063303-5
Itskos et al.
10 nm PFO film on GaN. The coefficient exhibits a modest
value of 0.15 at t ¼ 0, corresponding roughly to a 20% linear
polarized emission, and decays to a good approximation
mono-exponentially with a time constant of 500 ps. The anisotropy coefficient does not decay to zero saturating at a value
of 0.04. Time-integrated PL spectra show a steady-state anisotropy coefficient of 0.08, corresponding roughly to a linear
polarized emission of the 0-0 PFO vibronic of 12%. In the
same graph r(t) for a thicker 25 nm PFO film on GaN is also
displayed (orange curve). The anisotropy coefficient exhibits
a significantly smaller value of 0.04 at t ¼ 0 and decays to
zero within 400 ps, while time integrated spectra show a linear
polarization of less than 5%.
The maximum (t ¼ 0) value of the emission anisotropy
of all studied films is summarized at the bottom of Fig. 6. A
dependence of anisotropy on film thicknesses is evident on
both GaN and Al2O3 deposited films, with significant higher
values obtained for the ultrathin 10 and 15 nm films. Furthermore, the coefficient was found to decay slower for thinner
films and saturates at small but non-zero values for the thin
10 and 15 nm films while for thicker films it decays to zero.
In the geometry employed in our experiments, the data imply
a preferential in-plane orientation of the transition dipole
moments for thin films which in turn indicates that a fraction
of the polymer chains in such films acquire a more rigid, planar conformation. Similar trends of the optical anisotropy of
spin-coated films with thickness have been observed for
other conjugated polymers indicating a polymer chain alignment parallel to the substrate.27 In our case, the conformation
in which the fluorene repeat units are planarized with respect
to each other is known as b-phase.13,14,28 Such a phase is usually induced by thermal treatment or solvent annealing, however, it is possible that at the very small thicknesses employed
even unprocessed spin-casted films contain a fraction of bchains. The assignment is supported by the findings of ellipsometry and time-resolved measurements that show enhanced
optical constants and oscillator strength of the 0-0 vibronic
which is a typical optical signature, among others, of the PFO
b-phase. It is noted that non-zero but significant smaller emission anisotropy was measured on films of the same thickness
on fused silica (amorphous) substrates indicating that the crystallinity and polarizability of the GaN and Al2O3 substrates
affects the conformation of the overlaying PFO chains.
In summary, the PFO/GaN interface was investigated by
photoemission experiments to obtain the energy band diagram
of the two materials. Supported by PL studies that show evidence of photo-induced electron transfer from the PFO to
GaN, the measurements indicate that the GaN/PFO interface
can be exploited in sensor devices in which light harvesting
occurs on the highly absorbing polyfluorene material with
subsequent interfacial exciton dissociation and electron/hole
transport on the inorganic/organic component, respectively.
We have also investigated the effect of the thickness on the
photophysics of PFO films deposited on crystalline (GaN,
Al2O3) and amorphous (fused silica) substrates. At ultrathin
films on the crystalline substrates, enhanced oscillator
strengths and a preferential ordered planar chain conformation, most probably attributed to the b-phase of PFO, was
found to exist. Such findings have implications to light emitting structures employing ultrathin polyfluorene films in
Appl. Phys. Lett. 102, 063303 (2013)
nitrides such as those reported in FRET-coupled InGaN-polyfluorene hybrids.8,15–17 A preferential alignment of the optical
dipole moments of the polymer may result in an increased
dipole-dipole interaction. The hypothesis is supported by a
recent theoretical study29 that shows that the FRET rate can
be enhanced by up to a factor of two depending on the crystal
orientation of the organic material.
G. Itskos, X. Xristodoulou, E. Iliopoulos, M. Neophytou
and S. Choulis thank the Cyprus Research Promotion Foundation for financial support of the project via the infrastructure upgrade research Grant “ANABAHMIRH/0609/15.”
1
V. M. Agranovich, Yu. N. Gartstein, and M. Litinskaya, Chem. Rev. 111,
5179 (2011).
2
F. Hide, P. Kozodoy, S. P. DenBaars, and A. J. Heeger, Appl. Phys. Lett.
70, 2664 (1997).
3
J. O. McCaldin, Prog. Solid State Chem. 26, 241 (1998).
4
H.-F. Xiang, S.-C. Yu, C.-M. Che, and P. T. Lai, Appl. Phys. Lett. 83,
1518 (2003).
5
H.-J. Kim, J.-Y. Jin, Y.-S. Lee, S.-H. Lee, and C.-H. Hong, Chem. Phys.
Lett. 431, 341 (2006).
6
I. O. Huyal, U. Koldemir, T. Ozel, H. V. Demir, and D. Tuncel, J. Mater.
Chem. 18, 3568 (2008).
7
G. Heliotis, P. N. Stavrinou, D. D. C. Bradley, E. Gu, C. Griffin, C. W.
Jeon, and M. D. Dawson, Appl. Phys. Lett. 87, 103505 (2005).
8
C. R. Belton, G. Itskos, G. Heliotis, P. N. Stavrinou, P. G. Lagoudakis,
J. Lupton, S. Pereira, E. Gu, C. Griffin, B. Guilhabert, I. M. Watson, A. R.
Mackintosh, R. A. Pethrick, J. Feldmann, R. Murray, M. D. Dawson, and
D. D. C. Bradley, J. Phys. D: Appl. Phys. 41, 094006 (2008).
9
Y. Yang, G. A. Turnbull, and I. D. W. Samuel, Appl. Phys. Lett. 92,
163306 (2008).
10
H. Kim, Q. Zhang, Yoon-Kyu Song, A. Nurmikko, Q. Sun, and J. Han,
Phys. Status Solidi C 6, 593 (2009).
11
H. Kim, Ze-Lei Guan, Q. Sun, A. Kahn, J. Han, and A. Nurmikko, J. Appl.
Phys. 107, 113707 (2010).
12
N. Matsuki, Y. Irokawa, Y. Nakano, and M. Sumiya, Sol. Energy Mater
Sol. Cells 95, 284 (2011).
13
U. Scherf and E. J. W. List, Adv. Mater. 14, 477 (2002).
14
R. Abbel, A. P. H. J. Schenning, and E. W. Meijer, J. Polym. Sci. A:
Polym. Chem. 47, 4215 (2009).
15
G. Heliotis, G. Itskos, R. Murray, M. D. Dawson, I. M. Watson, and D. D. C.
Bradley, Adv. Mater. 18, 334 (2006).
16
G. Itskos, G. Heliotis, P. G. Lagoudakis, J. Lupton, N. P. Barradas, E.
Alves, S. Pereira, I. M. Watson, M. D. Dawson, J. Feldmann, R. Murray,
and D. D. C. Bradley, Phys. Rev. B 76, 035344 (2007).
17
G. Itskos, C. R. Belton, G. Heliotis, I. M. Watson, M. D. Dawson, R.
Murray, and D. D. C. Bradley, Nanotechnology 20, 275207 (2009).
18
A. Othonos, G. Itskos, D. D. C. Bradley, M. D. Dawson, and I. M. Watson,
Appl. Phys. Lett. 94, 203102 (2009).
19
C. I. Wu, A. Kahn, N. Taskar, D. Dorman, and D. Gallagher, J. Appl.
Phys. 83, 4249 (1998).
20
T. E. Cook, Jr., C. C. Fulton, W. J. Mecouch, K. M. Tracy, R. F. Davis, E. H.
Hurt, G. Lucovsky, and R. J. Nemanich, J. Appl. Phys. 93, 3995 (2003).
21
L. S. Liao, M. K. Fung, L. F. Cheng, C. S. Lee, S. T. Lee, M. Inbasekaran,
E. P. Woo, and W. W. Wu, Appl. Phys. Lett. 77, 3191 (2000).
22
S. F. Alvarado, P. F. Seidler, D. G. Lidzey, and D. D. C. Bradley, Phys.
Rev. Lett. 81, 1082 (1998).
23
J. Hwang and A. Khan, J. Appl. Phys. 97, 103705 (2005)
24
M. Stoessel, G. Wittmann, J. Staudigel, F. Steuber, J. Bl€assing, W. Roth,
H. Klausmann, W. Rogler, J. Simmerer, A. Winnacker, M. Inbasekaran,
and E. P. Woo, J. Appl. Phys. 87, 4467 (2000).
25
B. Johs, C. M. Herzinger, J. H. Dinan, A. Cornfeld, and J. D. Benson, Thin
Solid Films 313–314, 137 (1998).
26
M. Campoy-Quiles, G. Heliotis, R. Xia, M. Ariu, M. Pintani, P. Etchegoin,
and D. D. C. Bradley, Adv. Funct. Mater. 15, 925 (2005).
27
U. Zhokhavets, G. Gobsch, H. Hoppe, and N. S. Sariciftci, Thin Solid
Films 451–452, 69 (2004).
28
A. J. Cadby, P. A. Lane, H. Mellor, S. J. Martin, M. Grell, C. Giebeler,
D. D. C. Bradley, M. Wohlgenannt, C. An, and Z. V. Vardeny, Phys. Rev. B
62, 15604 (2000).
29
S. Kawka and G. C. La Rocca, Phys. Rev. B 85, 115305 (2012).
Downloaded 07 Oct 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions